learning_rate: 5e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant_with_warmup
lr_scheduler_warmup_steps: 50
training_steps: 500
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Global Wer
0.5587	0.0048	50	0.4012	61.1594
0.3825	0.0095	100	0.3882	40.3252
0.3474	0.0143	150	0.3583	33.2636
0.3047	0.0191	200	0.3598	35.1066
0.3975	0.0238	250	0.3386	40.6711
0.3738	0.0286	300	0.3330	45.0969
0.4058	0.0333	350	0.3264	53.1317
0.3196	0.0381	400	0.3181	33.0833
0.3361	0.0429	450	0.3144	33.1074
0.3953	0.0476	500	0.3098	33.5139

dianavdavidson/wh_l_v3_t_mucs_30_70_mucs_52473__val_finetune_trial

wh_l_v3_t_mucs_30_70_mucs_52473__val_finetune_trial